Professional Sovereignty Revisited: The Network Transformation of American Medicine?

A quarter century ago, the Pulitzer Prize celebrated two monumental works that chronicled and analyzed major aspects of the evolution of organizational form in the United States. Alfred Chandler's The Visible Hand (1977) told the story of the rise of the large managerial corporation that had dominated the landscape of industrial production for a century. And Paul Starr's The Social Transformation of American Medicine (1982) gave an account of the evolution of the medical profession and of the development of what is arguably America's most important "service" sector. Whereas Chandler told of how "rationalized" vertically integrated production had won out over the decentralized production and trade of the antebellum period, Starr described how decentralized production by the medical profession actually forestalled a corporate rationalization of medical care.

Both books influenced importantly how historians, sociologists, and perhaps even a few economists understood American organizational history. Both authors share a presumption—a centralist bias—about how organizational form related to what is rational and efficient. Chandler saw his vision realized in the modern managerial corporation; Starr saw the Chandlerian vision thwarted by a decentralized system of production that wouldn't go away. For Starr, this represented the triumph of professional authority over society's aspirations for a more efficient health care system.

The year 2004 is not 1977 or even 1982. In the years since these books appeared, American manufacturing has undergone an astounding transformation away from the model Chandler championed and that Starr viewed as inevitable if not entirely welcome. Neither could imagine that a decentralized system of production with low levels of vertical integration could not only triumph but also represent an increasingly effective and well-adapted modern mode of production.

The historical record shows that, in the quarter century since Starr wrote, levels of vertical integration have declined throughout the economy, as has the significance of managerial hierarchies in production. In their place a panoply of hybrid and network forms of organization has risen to prominence. These, while often maintaining and mixing some degree of hierarchy, are primarily oriented to smaller-scale relational contracting and market exchange (Lamoreaux, Raff, and Temin 2003). These smaller-scale forms have always been with us and have long played a significant role in economic growth (Scranton 1997). And it may well be that, although functional in its time and place, the vertically integrated managerial corporation—not decentralized production—is the historical anomaly (Langlois 2003).

Business historians today are reinterpreting Chandler's account in light of the facts of the past quarter century: it is not so much that Chandler's account is wrong as that it needs to be placed in a context wide enough to encompass as effective many other forms of economic organization, including markets and various hybrid forms, like networks. This essay approaches Starr in the same spirit. I will argue that, far from being obstructions to efficient production, networks of autonomous agents are useful mechanisms for managing a complex and contingent production problem. Taking this perspective throws Starr's magisterial history into a new light and suggests a way of explaining the recent developments in American medicine on which Starr speculated.

The Corporate Counterfactual

As Starr (1982: 7) rightly asserts, "the problem of professional sovereignty in American medicine is historical; . . . social structure is the outcome of historical processes. To understand a given structural arrangement, like professional sovereignty, one has to identify the ways in which people acted, pursuing their interests and ideals under definite conditions, to bring that structure into existence." But Starr is not concerned only with what people did; he is concerned also with what they might have done. It is Starr's first premise that things might have turned out differently. And he is quite clear that things ought to have turned out differently. Implicitly and sometimes explicitly, Starr is therefore carrying forward a counterfactual along with his chronicle.¹

Although Starr has little theory behind his counterfactual, he does have an unargued presupposition. He seems to take it as essentially self-evident that rationalized hierarchical organization is superior to other forms under all...